The overall aim of the research is to determine the composition (especially the major monomeric units and their derivatives) and physical-chemical properties of neuromelanin (NM) and then to relate these properties to possible roles in pathophysiology, including the effects of the alteration of NM in aging and in degenerative diseases, especially Parkinson's Disease. Since NMs arise from different precursors than eumelanins, their EPR spectra should show subtle but crucially important and useful differences when examined at different pH's and/or very high frequency (VHF). By increasing the magnetic field and RS frequency by a factor of ten (3.3 T and 94 GHz), the splitting between features having different g-factors (or chemical shifts) will also increase by a factor of ten. In the case of free-radicals such splittings often are too small to resolve with conventional EPR, and the increase in resolution at 94 GHz is quite dramatic. The g-factor of planar pi-radicals is determined prima rily by the spin density on heteroatoms such as O, N, and S, since they have spin-orbit coupling constants much larger than that for carbon. Thus, one can readily distinguish between semiquinone, semiquinone imine, and hydroxybenzothiazine radicals from their g-factors. W-band spectra of human neuromelanin samples show marked differences as a function of pH and differences between spectra of model melanins. It should be stressed that no such consistent differences can be detected at conventional EPR frequencies (X-band). This is an ongoing project.